RSP – EcoAssist

The focus of the research project on the optimization of life cycle analyses (LCA) is on the development of a data-based and technologically sound system that enables precise analyses. The first step involves the systematic collection and structuring of relevant emissions data from various sources such as databases, reports and scientific articles. This data is converted into standardized formats and visualized in a relational data model to ensure consistency and traceability. Based on this, specific requirements and a set of rules are formulated that serve as a methodological basis for further development.

To support the analysis and modelling processes, a modular toolkit is being developed that enables data-driven tasks such as automation and visualization. In addition, machine-learning algorithms are being developed that analyze complex data structures and iteratively improve data acquisition logics. These logics are evaluated both theoretically and simulatively in order to test their accuracy and robustness under realistic conditions.

Open source tools for modelling and visualization are used to present the results transparently. At the same time, weighting-based algorithms are being developed to enable data-based decisions and derive measures to optimize environmental impacts. The integration of a vector database and large language models (LLMs) forms the basis for a scalable and powerful system that supports both knowledge retrieval and complex data analyses.

The final step is the verification of the system through extensive tests that ensure functionality and user-friendliness. Detailed project documentation guarantees the traceability of the methods and results and provides a solid basis for future applications and further developments.